You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@ignite.apache.org by "Konstantin Orlov (Jira)" <ji...@apache.org> on 2023/02/24 14:51:00 UTC

[jira] [Created] (IGNITE-18902) Too much threads started on empty cluster

Konstantin Orlov created IGNITE-18902:
-----------------------------------------

             Summary: Too much threads started on empty cluster
                 Key: IGNITE-18902
                 URL: https://issues.apache.org/jira/browse/IGNITE-18902
             Project: Ignite
          Issue Type: Improvement
            Reporter: Konstantin Orlov
         Attachments: after_start.txt, after_table_creation.txt

Seems we start unreasonable amount of threads. Thread dump right after the start of a single node shows a 170 threads with prefix <node_name> ('idt_n_0' in dumps), 157 of them belongs to the JRaft. Creation of a table contributes another 160 threads to the dump (330 in total), and 114 of them belongs to the JRaft (271 in total).

Let's investigate if we really need all those threads or we can do better. You could find thread dumps attached below.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)